Fast reversed-phase liquid chromatography to reduce back exchange and increase throughput in H/D exchange monitored by FT-ICR mass spectrometry

Abstract

In solution-phase hydrogen/deuterium exchange (HDX), it is essential to minimize the back-exchange level of H for D after the exchange has been quenched, to accurately assign protein conformation and protein-protein or protein-ligand interactions. Reversed-phase HPLC is conducted at low pH and low temperature to desalt and separate proteolytic fragments. However, back exchange averages roughly 30% because of the long exposure to H 2 O in the mobile phase. In this report, we first show that there is no significant backbone amide hydrogen back exchange during quench and digestion; backbone exchange occurs primarily during subsequent liquid chromatography separation. We then show that a rapid reversed-phase separation reduces back exchange for HDX by at least 25%, resulting from the dramatically reduced retention time of the peptide fragments on the column. The influence of retention time on back exchange was also evaluated. The rapid separation coupled with high-resolution FT-ICR MS at 14.5 T provides high amino acid sequence coverage, high sample throughput, and high reproducibility and reliability. olution-phase hydrogen/deuterium exchange (HDX) coupled with mass spectrometry is a non-perturbative method to study protein conformational changes in protein-ligand and protein-protein interactions [1,2]. The exchange reaction rate minimizes at pH 2.0 -3.0 and at a temperature of about 0°C. To minimize back exchange of amide backbone -ND to -NH, the procedures after quench are performed under quench conditions. However, back exchange still averages nearly 30% [3], depending on various factors, e.g., peptide sequence [4]. With matrix-assisted laser desorption/ionization, the back-exchange level is even higher [5].In addition to low pH and low temperature, the H 2 O concentration should be kept low and exposure to H 2 O should be minimized during and after quench. Thus, elution of peptide fragments should occur as fast as possible. Our standard reversedphase liquid chromatography (RPLC) C 5 column elutes peptide fragments in about 4.5 min. Supercritical fluid chromatography (SFC) reduces back exchange because its mobile phase is mainly CO 2 instead of H 2 O [6]. Dramatic reduction of back exchange has been achieved with SFC separation. However, short column life and reliability of the experiment still require improvement.Here we present results with a high flow rate reversedphase ProZap™ C 18 column to reduce back exchange, while also increasing sequence coverage and total fragment number. Compared to a conventional C 5 column, the ProZap C 18 column has a wider inside diameter, shorter length, smaller particle size, and larger pore size, enabling higher flow rate and lower back pressure. The retention time is thereby reduced from 4.5 min (Jupiter™ C 5 column) to 1.5 min, for a 3-fold reduction in H 2 O exposure time and concomitant reduction in back exchange for the fragment peptides. Besides reduction in back exchange, the ProZap column also provides much improved elution efficiency for increased fragment number and...

‘Fixed charge’ chemical derivatization and data dependant multistage tandem mass spectrometry for mapping protein surface residue accessibility

“…It can be seen that both these MS/MS spectra are identical, each giving rise to a single product ion via the exclusive loss of dimethylsulfide (S(CH 3 ) 2 ). Similar results were obtained by dissociation of the doubly charged ([1-13 8 ϩH] 2ϩ and [1-13 1 ϩH] 2ϩ ) precursor ions (shown in the insets to Figure 2a and b). Notably, the presence of a 'mobile' proton in the triply charged precursor ions (the doubly charged precursor ions are in a 'non-mobile' charge state) was observed to have no effect on the MS/MS fragmentation reaction involving the exclusive loss of S(CH 3 ) 2 , consistent with previous studies demonstrating that the selective dissociation of sulfonium ion derivatized peptides occurs independently of the proton mobility of the precursor ion, i.e., that ionizing protons simply act as 'spectators' rather than participating in competing fragmentation reactions [28 -33].…”

“…Figure 2 shows the CID-MS/MS and -MS 3 spectra acquired from peptide 1-13 (PNFSGNWKIIRSE), which contains two modifiable residues, i.e., the secondary amino group of the N-terminal proline residue and the -amino of the lysine residue K 8 . Thus, it is expected that two singly modified peptides (i.e., 1-13 1 (where the superscript number indicates the modified residue) and 1-13 8 , with the same mass but different modification sites), and one doubly modified peptide (1-13 (1,8) should be observed. Figure 2a and Figure 1.…”

“…However, as the abundance of the doubly charged precursor ions were significantly lower than those for the triply charged precursors, the signal to noise for the MS 3 spectra were substantially lower. Figure 3a and b show the CID-MS/MS product ion spectra obtained from the doubly modified (1-13 1,8 peptide, in its doubly (i.e., [1-13 1,8 ] 2ϩ ) and triply (i.e., [1-13 1,8 ϩH] 3ϩ ) charged precursor ion charge states, respectively. In each case, two characteristic neutral loss product ions were observed, corresponding to the neutral losses of one and two S(CH 3 ) 2 groups, indicative of the presence of two modifications within the peptide.…”

“…This value is 17.2 kcal mol Ϫ1 higher than that calculated for the simplest model for lysine (methylamine, 211.0 kcal mol Ϫ1 ), and is intermediate between the calculated values for the simplest models for histidine (4-methyl imidazole, 223.5 kcal mol Ϫ1 ) and arginine (N-methyl guanidine, 234.2 kcal mol Ϫ1 ), at the same level of theory. Despite this result, a comparison of the MS 3 product ion spectra obtained from the modified 1-13 peptide ions (e.g., the [1-13 8 Figure 3c, with the MS/MS spectra of the unmodified 1-13 peptide in the same charge state (i.e., the [1-13ϩ3H] 3ϩ ion) (Supplementary Figure S1), revealed that the sequence coverage obtained following MS 3 of the modified peptide ions was equivalent to, or greater than, that from the unmodified MS/MS spectrum. Interestingly, the MS 3 spectrum obtained from the modified [1-13 8 -S(CH 3 ) 2 ϩ2H] 3ϩ product ion (K 8 modified) was very similar to that obtained by MS/MS of the unmodified [1-13ϩ3H] 3ϩ ion.…”

“…HDX may potentially be used to probe the entire protein backbone, thereby providing the highest resolution of the MS-based approaches. However, limitations of HDX that can hamper determination of the precise location of exchange sites and therefore limit the spatial resolution that can be obtained, are the back-exchange of deuterium to hydrogen during proteolysis and sample handling before MS analysis [7,8], and hydrogen/deuterium 'scrambling' during CID-MS/MS analysis [9]. Recently, the latter problem has been largely overcome by the use of alternate dissociation techniques such as electron capture dissociation (ECD) [10] and electron-transfer dissociation (ETD) during MS/MS [11].…”